3 Acid Stress Survival Mechanisms of the Cariogenic Bacterium Streptococcus mutans

Streptococcus mutans, the major etiological agent in human dental caries, is capable of forming a biofilm, or dental plaque, on the tooth surface (Loesche, 1986; Tanzer et al., 2001). S. mutans generates large amounts of acid within dental plaque from fermentable dietary carbohydrates. During meals, the ingestion of carbohydrates causes the pH of the dental plaque to fall below 4.0. Acid accumulation can eventually destroy the crystalline structure of teeth that is the hardest tissue in the human body, leading to the formation of a carious lesion (Quivey et al., 2001). The ability of S. mutans to survive in such a severe environment represents one of the most important virulence factors of this microorganism. The mechanisms of acid tolerance that are most common among Gram-positive bacteria have been proposed to be: i) proton pumps; ii) protection and/or repair of macromolecules; iii) cell-membrane changes; iv) production of alkali; v) regulators; vi) cell density and biofilms; and vii) alteration of metabolic pathways (Fig. 1) (Cotter & Hill, 2003). Many researchers have sought to explain the mechanisms of acid tolerance in S. mutans, and various genes contributing to aciduricity in S. mutans have been identified. In this chapter, we review those genes that have been reported to be involved in S. mutans aciduricity, including those participating in two-component systems and others, especially targeting the dgk homolog.


Introduction
Streptococcus mutans, the major etiological agent in human dental caries, is capable of forming a biofilm, or dental plaque, on the tooth surface (Loesche, 1986;Tanzer et al., 2001).S. mutans generates large amounts of acid within dental plaque from fermentable dietary carbohydrates.During meals, the ingestion of carbohydrates causes the pH of the dental plaque to fall below 4.0.Acid accumulation can eventually destroy the crystalline structure of teeth that is the hardest tissue in the human body, leading to the formation of a carious lesion (Quivey et al., 2001).The ability of S. mutans to survive in such a severe environment represents one of the most important virulence factors of this microorganism.The mechanisms of acid tolerance that are most common among Gram-positive bacteria have been proposed to be: i) proton pumps; ii) protection and/or repair of macromolecules; iii) cell-membrane changes; iv) production of alkali; v) regulators; vi) cell density and biofilms; and vii) alteration of metabolic pathways (Fig. 1) (Cotter & Hill, 2003).Many researchers have sought to explain the mechanisms of acid tolerance in S. mutans, and various genes contributing to aciduricity in S. mutans have been identified.In this chapter, we review those g e n e s t h a t h a v e b e e n r e p o r t e d t o b e i n v o l v e d i n S. mutans aciduricity, including those participating in two-component systems and others, especially targeting the dgk homolog.

Two-component system
Two-component systems (TCSs), prokaryote-specific signal transduction systems, are widespread in prokaryotes and play extensive roles in adaptation to environmental changes.The TCS operon (tcs) consists of hk, which encodes a sensory histidine kinase (HK), and rr, which encodes its cognate response regulator (RR).The HK undergoes autophosphorylation on a histidine residue in response to a specific environmental signal and relays this phosphate group to an aspartic acid residue on the cognate RR.The phosphorylated RR then binds target DNA elements with greater affinity, inducing or repressing the transcription of target genes (Hoch, 2000;Rampersaud et al., 1994).In this way, bacteria are able to adapt to the changes in external environment and to modulate gene expression.TCSs may be responsible for the acid tolerance of S. mutans.Analysis of the complete genome sequence of S. mutans UA159 suggested the presence of 13 hk-rr homologs and one orphan rr homolog (Table 1, smtcs02-15) (Ajdic et al., 2002).The roles of some specific tcs genes in acid tolerance have been evaluated.Li et al. showed that disruption of smhk13 or smrr13 resulted in a diminished log-phase acid-tolerance response in S. mutans BM71 (Li et al., 2001) and that only smhk02 of smtcs02 was involved in the acidtolerance response of S. mutans NG8 (Li et al., 2002).Qi et al. (2004) andAhn et al. (2006) reported that the smhk08 mutant exhibited a significant growth defect, whereas the growth of both the smrr08 mutant and smhk-rr08 double mutant was similar to that of wild-type UA159 when grown at pH 6.4.Ahn et al. (2006) also showed that all smhk08, smrr08, or smhk-rr08 mutants presented growth defects when grown at pH 5.5.Then, Lévesque et al. (2007) systematically inactivated each of the 13 hk, but not rr, genes in S. mutans UA159 and evaluated the roles of the hk genes in acid tolerance.They showed that smhk09 and smhk14 were involved in S. mutans acid tolerance.Furthermore, Biswas et al. (2007) found an additional tcs (Table 1, smtcs01) in the genome of S. mutans UA159 and examined the involvement of 14 hk genes in acid tolerance.They showed that only smhk08 was involved in aciduricity.However, these studies focused only on the role of HKs and so did not provide a comprehensive overview of the role of TCSs in acid tolerance.Therefore, we systematically constructed rr deletion mutants and hk-rr double mutants of S. mutans UA159 and examined the effect on acid tolerance (Kawada-Matsuo et al., 2009).
Thirteen rr mutants and twelve hk-rr double mutants were obtained, the exceptions being smrr10, smtcs10, smrr12, and smtcs12.The derivation of null mutations of these genes was unsuccessful, probably due to a loss of viability of these mutants.To examine the effects of these rr mutations on the acid tolerance of S. mutans, wild-type UA159 and the 25 mutants were grown in brain-heart infusion (BHI) broth adjusted to pH 7.2 or pH 5.5.Growth curves were generated, and the mid-log-phase doubling time was determined.All rr and hk-rr mutants grew similarly to wild-type UA159 at pH 7.2.However, as shown in Table 2, deletion of four rr genes (smrr03, smrr05, smrr08, and smrr13) caused significantly decreased growth rates compared with that of wild-type UA159 when grown at pH 5.5.The growth rates of the hk-rr double mutants were similar to those of the corresponding rr mutants, and the differences in doubling time between them were not significant.The finding that smrr08 and smrr13 were involved in the acid tolerance of S. mutans is consistent with previous findings (Ahn et al., 2006;Li et al., 2001).On the other hand, smrr03 and smrr05 were, for the first time, demonstrated to be involved in S. mutans UA159 acid tolerance.However, Lévesque et al. (2007) and Biswas et al. (2007) showed that inactivation of their cognate hk genes did not affect acid tolerance.To confirm whether only the rr gene is involved in acid tolerance in the smtcs03 and smtcs05 mutants, the hk genes of smtcs03 and smtcs05 were individually inactivated, and the acid tolerance ability of these mutants compared.As shown in Table 3, the smhk03 mutant exhibited a decreased growth rate compared with the wild type when grown at pH 5.5.This was not consistent with previous results.In contrast, the smhk05 mutant grew similarly to wild-type UA159 at pH 5.
where k is the growth rate, and g = 0.693/k, where g is the doubling time.Values are the mean ± standard deviation obtained from three independent experiments.b Wild-type strain c Orphan rr * Significant increase from Td of wild-type UA159 by Tukey's HSD, p < 0.05 ** Significant increase from Td of wild-type UA159 by Tukey's HSD, p < 0.01 Table 2. Doubling times of rr or hk-rr deletion mutants at pH 5.5.

www.intechopen.com
Generally, a signal sensed by a HK is thought to be transmitted to the cognate RR via transfer of phosphoryl groups, and deletion of either the hk or rr should generate a similar phenotype.However, we found that only the rr of smtcs05 was involved in S. mutans acid tolerance.Furthermore, as mentioned above, Li et al. (2002) reported that the hk, but not the rr, of smtcs02 was involved in the acid tolerance of S. mutans NG8.Qi et al. (2004) and Ahn et al. (2006) also reported that the smhk08 mutant showed a significant growth defect, whereas the growth of both the smrr08 mutant and smhk-rr08 double mutant was similar to that of wild-type UA159.

Genes other than TCS involved in S. mutans acid tolerance
Many studies have implicated genes other than those involved in TCS in S. mutans aciduricity.Table 4 summarizes the characteristics of 14 such genes, of which the functions of the products of only nine have been experimentally verified.These are aguA, encoding an agmatine deiminase that is involved in alkali production (Griswold et al., 2004); dltC, involved in the synthesis of D-alanyl-lipoteichoic acid, which is associated with alteration of membrane composition (Boyd et al., 2000); gluA, encoding a glucose-l-phosphate uridylyltransferase involved in the synthesis of UDP-D-glucose (Yamashita et al., 1998); lgl, encoding a lactoylglutathione lyase involved in the detoxification of methylglyoxal (Korithoski et al., 2007); luxS, encoding a Sribosylhomocysteine lyase involved in autoinducer AI2 synthesis (Wen and Burne, 2004); and uvrA, encoding an excinuclease ABC subunit A that is involved in DNA repair (Hanna et al., 2001).Mutation of three genes (ffh, ftsY, yidC2 genes) that are involved in the signal recognition particle pathway, significantly reduced H + /ATPase specific activity compared with that of the wild type (Hasona et al., 2005).However, how these functions contribute to S. mutans aciduricity remains unclear.Furthermore, the functions of the five remaining gene products were predicted based on DNA sequence homology, and so estimation of their role in S. mutans aciduricity is much more difficult.To examine the extent to which a gene contributes to S. mutans acid tolerance, 14 mutants in which one of the genes listed in Table 4 was inactivated and the dgk mutant were constructed from S. mutans UA159, and their growth at pH 7.4 and at pH 5.5 was compared with that of the wild type (Fig. 2) (Shibata et al., 2011).Inactivation of aguA, brpA, glrA, htrA, lgl, luxS, ropA, or uvrA did not significantly affect the acid tolerance of S. mutans as compared with wild-type UA159 when grown at pH 5.5.Three types of method of assessing the acid tolerance of S. mutans are available.Simple acid tolerance is evaluated by a procedure in which over-night cultures or log-phase cells grown at neutral pH are subcultured in media or on plates at neutral and acidic pH.Another method is acid killing, in which log-phase cells grown at neutral pH are incubated at a lethally acidic pH, and then viability is determined by plate counts.The final method is determination of the acid tolerance response in which viability is estimated by plating after incubation at neutral or acidic pH of log-phase cells grown at neutral pH followed by incubation at killing pH.We used the first method, and so any discrepancy between this and other studies may derive from differences in the method used.However, comparing all of the mutants in terms of the most basic characteristics of acid tolerance is a worthwhile endeavor.Of course, a comparison using the other criteria is important, and will be performed as part of our next effort.Notably, the dgk and the gluA mutants grew extremely slowly at pH 5.5, although the clpP, dltC, ffh, ftsY, and yidC2 mutants also displayed significant reductions in growth rate at pH 5.5 compared with the wild-type UA159.However, only the brpA, dgk, dltC, htrA, lgl, luxS, ropA, and uvrA mutants showed growth rates comparable to the wild-type strain at pH 7.4.These findings suggest that the reduction in growth rates of the clpP, ffh, ftsY, yidC2, and gluA mutants at acidic pH values might be derived from reduced viability and not specifically related to acid tolerance.Therefore, dltC and dgk are likely specifically involved in acid tolerance.Of these two genes, the striking reduction in growth rate of the dgk mutant at acidic pH indicates that dgk is of great interest for elucidating the acid-tolerance mechanism of S. mutans.

Diacylglycerol kinase
Diacylglycerol kinase (Dgk) catalyses the ATP-dependent phosphorylation of sn-1,2diacylglycerols, resulting in production of phosphatidic acid.In eukaryotic cells, diacylglycerol and phosphatidic acid are immediate second cellular messengers responding to extracellular signals, suggesting that Dgk is a key enzyme in cellular signal transduction (Moolenaar et al., 1986;Murayama & Ui, 1987;Nishizuka, 1984;Topham & Prescott, 1999).Among bacterial Dgk, only that of Escherichia coli has been well-characterized, and it is a small integral membrane protein with a molecular mass of 13.2 kDa.This enzyme functions in the recycling of the diacylglycerol produced during turnover of membrane phospholipids (Hasin & Kennedy, 1982;Rotering & Raetz, 1983) and plays an important physiological role in responding to environmental stress as well as its role in eukaryotic cells (Raetz & Newman, 1979;Walsh et al., 1986).On the other hand, the S. mutans Dgk homolog has a molecular mass of 15.3 kDa and comprises 137 amino acids.It is interesting that insertion of the transposon Tn916 into the codon for the tenth amino acid from the C terminus of the Dgk homolog resulted in defective growth of the mutant (GS5Tn1) at acidic pH values (Yamashita et al., 1993).In addition to attenuation of aciduricity, this mutant possessed reduced resistance to high osmolarity and temperature (Yamashita et al., 1993).The C terminus of the Dgk homolog may thus play an important role in signal transduction during environment stress.To evaluate how the C terminus of Dgk contributes to S. mutans acid tolerance, we sequentially truncated amino acids from the C terminus of Dgk and finally constructed 11 mutants termed UADGK0-10, expressing Dgk0-10 (Fig. 3) (Shibata et al., 2009).The mutants showed no significant difference in growth rate at neutral pH (doubling times: 53.8 to 61.6 min; Table 5).Most, with the exception of UADGK0 to UADGK2, showed a reduction in growth rate at pH 5.5 compared with the wild type (Table 5 and Fig. 4).UADGK3, in which three amino acid residues had been deleted from the C-terminus of Dgk, showed a slight reduction in growth rate.Subsequent deletion of amino acids from the C-terminus resulted in further reductions in growth rate at acidic pH.Indeed UADGK4, UADGK5, and UADGK6 had significantly increased doubling times (p < 0.05, p < 0.001, and p < 0.0001, respectively) compared with UADGK0.UADGK7, in which seven amino acid residues had been deleted, showed extremely limited growth in the first 9 hours.Further truncation of the C-terminus of Dgk (UADGK8 to UADGK10) resulted in no growth at pH 5.5.These results suggest that the C-terminal of the Dgk homolog is indispensable for its function in aciduricity of S. mutans.We further constructed two additional UA159 dgk mutants, UADGK11 and UADGK12 (Fig. 3) to evaluate the function of truncated Dgk.There were only negligible differences in the growth rates of these two mutants at pH 5.5, 5.8, or 6.3, compared with that of UADGK10.(Shibata et al., 2009) to help readers understand.Fig. 4. Growth curves of S. mutans UA159 dgk mutants grown in BHI medium at pH 5.5.Growth was defined as the increase in OD 550 , and was calculated by subtraction of OD 550 at the initiation of growth from that at the times indicated.Data represent the means of three independent experiments.All the mutants names were changed from those of previous paper (Shibata et al., 2009)  Differences in the doubling time between UADGK0 and UADGK1-10 were analyzed by Bonferroni multiple comparison test (*, p < 0.05; **, p < 0.001; ***, p < 0.0001).a Statistical analyses were not carried out because of too slow growth rates.
All the mutants names were changed from those of previous paper (Shibata et al., 2009) to help readers understand.
Table 5.Effect of low pH on growth of S. mutans UA159 dgk mutants.
We next constructed recombinant Dgk proteins corresponding to S. mutans strains UADGK10, UADGK7, UADGK5, and UADGK0 utilizing E. coli strains RZDGK10, RZDGK7, RZDGK5, and RZDGK0, respectively.The kinase activity in cell lysates of E. coli transformants was examined by an octyl glucoside mixed-micelle assay (Preiss et al., 1986), using undecaprenol as a substrate because of it has a higher substrate specificity for the S. mutans Dgk homolog compared with diacylglycerol (Lis & Kuramitsu, 2003).
As shown in Fig. 5A, whereas the full-size S. mutans Dgk protein expressed in RZDGK0 catalyzed a high level of phosphorylation of undecaprenol, the Dgk missing five amino acid residues from the C terminus expressed in RZDGK5 exhibited markedly reduced kinase activity.Furthermore, RZDGK7 (seven amino acids missing from the C terminus) exhibited much weaker kinase activity than did RZDGK5.The deletion of 10 C-terminal amino acid residues of Dgk in RZDGK10 resulted in a total lack of kinase activity.These differences were confirmed by quantitative analysis (Fig. 5B).These data indicate that the C-terminus of the S. mutans Dgk homolog plays an important role in kinase activity and may harbor residues required for catalysis.Alternatively, incorrect folding of the protein due to the missing C-terminal residues may cause loss of kinase activity.Therefore, its catalysis of undecaprenol phosphorylation is closely related to S. mutans acid tolerance.All the E.coli strains names were changed from those of previous paper (Shibata et al., 2009) to help readers understand.
Moreover, the importance of the C-terminal end of Dgk in S. mutans acid tolerance was examined in a specific pathogen-free animal model (Table 6).The dgk mutant strain clearly displayed a significant reduction in smooth-surface caries compared with the wild type (p < 0.005).In contrast, no significant difference in plaque extent was observed between the wild-type and dgk mutant strains.These results suggest that aciduricity regulated by the dgk gene product might play a critical role in S. mutans virulence.ND, Not determined.∆, 4 units at risk; ∆∆, 12 fissures at risk; ∆∆∆, 20 units at risk. a Significant difference between water control and UA159 (wt), p < 0.05.b Significant difference between UADGK1 (dgk) and water control, p < 0.05.c Significant difference between UA159 (wt) and UADGK1 (dgk), p < 0.05.The mutant name was changed from those of previous paper (Shibata et al., 2009) to help readers understand.
Table 6.Influence of dgk deletion on smooth-surface plaque extent, initial and advanced dentinal fissure lesions, smooth-surface caries, and colonization properties.

A potential target for anti-caries chemotherapy
Development of an effective anti-caries agent is the ultimate goal of our work.Considering the characteristics of known mutants, the Dgk homolog seems to be the most promising target for anti-caries agents.Dgks have been extensively studied in mammals, and several inhibitory compounds, e.g., R59022 and R59949 (Fig. 6), have been reported.In contrast, inhibitors of prokaryotic Dgk have not yet been elucidated.When first attempting to discover inhibitors of prokaryotic Dgk, we tested the effects of R59022 and R59949 on the growth of S. mutans (Shibata et al., 2011).Although neither R59022 nor R59949 influenced growth at pH 7.4, R59949, but not R59022, showed a significant inhibitory effect at acidic pH (Fig. 7).Inhibition by R59949 increased by 13, 29, 58, 68, and 78% at pH 5.4, 5.3, 5.2, 5.1, and 5.0, respectively (Fig. 7A).These findings were particularly interesting because R59022 and R59949 were used at concentrations of 100 μM and 25 μM, respectively, due to the limited solubility of R59949.Furthermore, we examined the inhibitory effects of R59022 and R59949 on the kinase activity of S. mutans Dgk.Neither R59022 nor R59949 inhibited kinase activity at pH 7.4; this is in agreement with their lack of effect on S. mutans growth at neutral pH.As mentioned above, R59949 significantly inhibited the growth of S. mutans at acidic pH values (below 5.4).When evaluating the effect of R59949 on enzyme activity at acidic pH, it is important to know the intracellular pH of S. mutans cells; the intracellular pH of S. mutans cells was 6.4 when cultured in broth at pH 5.2.Therefore, we determined the inhibitory effect of R59949 and R59022 on S. mutans Dgk kinase activity at pH 6.4.R59949, but not R59022, inhibited kinase activity with undecaprenol as a substrate by around 20% (Fig. 8).S. mutans Dgk is inherently different from mammalian Dgk in terms of its molecular size, molecular structure, and substrate specificity.However, it is interesting that R59949 inhibits the enzymatic activity of S. mutans Dgk even with undecaprenol as the substrate.
Additionally, the difference in inhibitory activity between R59949 and R59022 means that a comparison of their molecular structure may lead to discovery of further potent Dgk inhibitors specific for prokaryotic enzymes, that is, new anti-caries agents.

Conclusion
The reduction of environmental pH in dental plaque by the cariogenic microorganisms is important step in the development of dental caries.The cariogenic microorganisms should survive in a relentless environment produced by themselves in order to exhibit their maximum virulence.In this chapter, we described the acid tolerance characteristics of the cariogenic microorganism, S. mutans.TCSs seem to be the most suitable system for adaptation to environmental conditions.However, no TCS seems to be definitively responsible for S. mutans acid tolerance.At present, identification of TCS depends on gene homology searching, which may not identify all genes encoding TCS that contribute to S. mutans acid tolerance.We focus on dgk because it is the most promising contributor to S. mutans acid tolerance when assessed using a simple acid tolerance assay.Although the precise mechanism by which the gene product is involved in acid tolerance has not yet been elucidated, dgk is the only gene whose product has been definitively implicated in cariogenicity in an animal model.Furthermore, potential specific inhibitors of the gene product have been introduced.This fact may aid in development of next-generation anti-caries therapies based on the ability of this microorganism to adapt to environmental conditions.However, much detail of the acid tolerance mechanisms of S. mutans remains unknown, and so further study is required.

Fig. 1 .
Fig. 1.Acid tolerance mechanisms proposed for gram-positive bacteria.The figure is taken from Cotter & Hill (2003) with some modification.

Fig. 2 .
Fig.2.Growth curves of S. mutans UA159 and mutant strains grown in BHI medium at pH 7.4 ( ) or pH 5.5 ( ).Growth was defined as the increase in OD 550 , and was calculated by subtraction of OD 550 at the initiation of growth from that at the times indicated.Data represent the means of three independent experiments.Graphs of mutant strains also represent the growth curves of wild-type UA159 at pH 7.4 ( ) and pH 5.5 ( ) as controls.

Fig. 3 .
Fig. 3. Representation of the truncated Dgk proteins used.The deduced amino acid sequence of the dgk gene from S. mutans UA159 is presented.The terminal amino acid of the truncated Dgk expressed in each UA159 dgk mutant and each E. coli RZ transformant is indicated along the sequences by a curved arrow, which indicates that the sequence is deleted from the right up to this site.All the truncated Dgk proteins names were changed from those of previous paper(Shibata et al., 2009) to help readers understand.

Fig. 5 .
Fig. 5. Effect of deletion of the C-terminal tail of Dgk on undecaprenol kinase activity.(A) Comparison of undecaprenol kinase activity of the full-size Dgk and various C-terminally truncated forms of Dgk.The undecaprenol kinase activity in the lysates from E. coli RZ cells was determined using an octyl glucoside mixed-micelle assay.(B) Quantification analysis of the kinase assay.Quantification was carried out by normalization of radioactive bands in the kinase assay using the protein level.Vertical bars represent standard deviation.Differences in kinase activity between RZDGK0 and RZDGK5, RZDGK7, or RZDGK10 were analyzed by Student's t test (*, p < 0.0001).All the E.coli strains names were changed from those of previous paper(Shibata et al., 2009) to help readers understand.

Fig. 7 .
Fig. 7. Effect of R59949 (A) and R59022 (B) on the growth of S. mutans.Data represent the mean ± standard deviation.Differences in growth rate between cells cultured in the presence and absence of Dgk inhibitor were analyzed using Student's t test.*, p < 0.05; **, p < 0.0001.

Fig. 8 .
Fig. 8. Effect of R59949 on kinase activity with undecaprenol as a substrate.Data represent the mean ± standard deviation.Differences in kinase activity between cells cultured in the presence and absence of R59949 were analyzed by Student's t test.*, p < 0.05.

Table 1 .
The tcs genes identified in the S. mutans UA159 genome.
tcs, two-component system; hk, histidine kinase; rr, response regulator.a These results suggested involvement of several TCSs in S. mutans aciduricity via cross-talk between different TCS components.After all, among 15 TCSs, seven (Smtcs02, 03, 05, 08, 09, 13, and 14) appear to be involved in S. mutans aciduricity.Nevertheless, inactivation of no single TCS caused a complete loss of acid tolerance.Therefore, other TCSs that are definitively related to S. mutans acid tolerance may exist, but they cannot be identified by homology searching.

Table 3
. Doubling times of hk, rr, and hk-rr deletion mutants of smtcs03 and smtcs05 at pH 7.2 and pH 5.5.

Table 4 .
Genes reported to be involved in S. mutans acid tolerance. www.intechopen.com to help readers understand.